SMALL - espritmodel.com SMALL - Telemetry SMALL - Radio
Reply
Thread Tools
Old Dec 15, 2005, 05:09 PM
Registered User
Joined Jan 2005
98 Posts
FET failure charctaristics?

Does anyone have any documentation or sources they can reccomend dealing with FET failures, how and why and what happens when they fail? The state they often fail in, maybe current charts or something?
Subarubrat is offline Find More Posts by Subarubrat
Reply With Quote
Sign up now
to remove ads between posts
Old Dec 15, 2005, 06:09 PM
CamLight Systems
New York City, USA
Joined Oct 2003
1,172 Posts
Mine have always failed by my overheating the junction, causing it to short. But, the short doesn't last long as the source resistor (for current sensing) usually vaporizes instantly and opens the circuit.

I would check http://www.irf.com (Intl. Rectifier) for more info. They have quite a few Application Notes and White Papers dealing with FETs.
JohnMuchow is offline Find More Posts by JohnMuchow
Reply With Quote
Old Dec 15, 2005, 09:38 PM
Xtreme Power Systems
Lake Havasu, AZ
Joined Jun 2005
15,938 Posts
Almost every failure is related to the design. Typically, people do not drive the gate with enough voltage (not current) to fully saturate the gate. Quite often, the input voltage is tied directly to the gate (P-Channel) and driven low to pass the source to the drain. The problem is that the gate requires a constant voltage and as batteries drop, the gate voltage drops below the required Vgs and is no longer fully saturated. This results in heat (extreme), causing failures. Other sources of failures include over-driving the gate with too much voltage, having too high of input voltage, and having too slow of switching speed between Vss and Vgs. As long as you stay within the limits of the part, and properly saturate the gate, a FET should never fail.
JimDrew is offline Find More Posts by JimDrew
RCG Plus Member
Last edited by JimDrew; Dec 15, 2005 at 09:46 PM.
Reply With Quote
Old Dec 16, 2005, 11:53 PM
Cat lover
Joined Mar 2005
68 Posts
In addition to what Jim and John have said, I'd like to chime in and mention something i've experienced with slow failing fets. When a FET is starting to fail for thermal reasons or whatever, its resistance increases. If you then reduce the load, you may never even realise the fet is half dead. But once you try to put a lot of current through it again, the increased resistance will result in a LOT more heat being generated in the fet, until it eventually completely fries.
Afroman is offline Find More Posts by Afroman
Reply With Quote
Old Dec 17, 2005, 06:15 AM
Registered User
vintage1's Avatar
East Anglia, UK
Joined Sep 2002
29,705 Posts
Thats right.

FETs fail in different ways to Biploars - there second breakdown and destrction of te vollector base junction is common.

With FETS you either get gate punch-through or severve modification of the doping levels and pattern in the channel.

Esenntially seminconductors are made with a semicinductor whose structure varies in space according to the dopant levels injected in those areas. Over time, the chemistry decays...at room temeperature te ageing is in decades, if not centuries..at temperatures about 200C its in hours, minurtes seconds or even less.

The FET gets 'old' very very quickly.

However its somewhat rare to see this - generally the FET is either working OK, or its dead.
vintage1 is offline Find More Posts by vintage1
Reply With Quote
Old Dec 17, 2005, 12:50 PM
แอนครู กรีน (เขียว)
C Andrew Green's Avatar
Oxford, UK
Joined Mar 2002
270 Posts
I was quite surprised when the SO8 (small 8 legged surface mount) FET failed in my MS (Jeti) brushed speed controller in a Hornet helicopter. I was flying around and thought that the batteries were sagging, when I landed and throttled back it just sat there revving because the poor little FET was done for! (short circuit source/drain) I held the model by the tailboom while a friend reached under and unplugged it (It's only a small heli). I replaced it with the equivalent part from International Rectifier and it has susequently flown loads more flights on 3 cel LiPos.

Andrew
C Andrew Green is offline Find More Posts by C Andrew Green
Reply With Quote
Reply


Thread Tools